Seasonality of primary productivity affects coastal species more than its magnitude

While the importance of extreme conditions is recognised, patterns in species' abundances are often interpreted through average environmental conditions within their distributional range. For marine species with pelagic larvae, temperature and phytoplankton concentration are key variables. Along the south coast of South Africa, conspicuous spatial patterns in recruitment rates and the abundances of different mussel species exist, with focal areas characterized by large populations. We studied 15 years of sea surface temperature (SST) and chlorophyll-a (chl-a) satellite data, using spectral analyses to partition their temporal variability over ecologically relevant time periods, including seasonal (101 to 365 days) and intra-seasonal cycles (20 to 100 days). Adult cover and mussel recruitment were measured at 10 sites along the south coast and regression models showed that about 70% of the variability in recruitment and adult cover was explained by seasonal variability in chl-a, while mean annual chl-a and SST only explained 30% of the recruitment, with no significant effect for adult cover. SST and chl-a at two upwelling centres showed less predictable seasonal cycles during the second half of the study period with a significant cooling trend during austral autumn, coinciding with one of the mussel reproductive peaks. This likely reflects recent changes in the Agulhas Current, the world's largest western boundary current, which affects coastal ecosystems by driving upwelling. Similar mechanisms probably operate in other marine systems with the potential to affect the distribution patterns of key ecosystem engineers. We propose that variability in the characteristic timescales of environmental fluctuations can explain the spatial patterns of abundance of foundational species by affecting larval recruitment. This is especially important in a context of global and pervasive climate change, as shifts in the periodicity of environmental fluctuations appear to reflect large scale climatic teleconnections driven by anthropogenic forcing.